Artificial intelligence is becoming part of everyday life for younger generations, helping with everything from schoolwork and productivity to creative projects and personal development. A new study published in JAMA Pediatrics reveals that AI is also becoming a source of emotional support for many teenagers and young adults.

According to research conducted by RAND, nearly one in five people aged 12 to 21 reported using AI chatbots such as ChatGPT, Google Gemini, or Character.AI for advice when feeling stressed, anxious, sad, or overwhelmed. The figure represents a significant increase compared to similar research conducted earlier in 2025, highlighting how quickly AI-powered conversations are becoming part of young people’s lives.

AI Is Filling a Growing Gap

The findings point to a reality many organizations are already observing: access to mental health resources remains limited, while demand continues to grow.

Researchers suggest that some young people may be using AI as an accessible alternative when professional support is unavailable, difficult to access, or too expensive. Others may simply feel comfortable engaging with AI because it is already integrated into many aspects of their daily routines.

Interestingly, most respondents reported finding the chatbot interactions helpful, and nearly two-thirds said they had never discussed their AI-assisted mental health conversations with anyone else.

From a technology perspective, this highlights a growing shift: AI systems are increasingly becoming trusted companions rather than simple productivity tools.

Where the Risks Begin

While conversational AI can provide information, reflection prompts, mindfulness suggestions, or general emotional support, experts warn that these systems were never designed to replace licensed mental health professionals.

Mental health crises, suicidal ideation, severe anxiety, trauma, and psychotic episodes require specialized human intervention, clinical expertise, and accountability structures that current AI systems cannot provide.

Researchers and healthcare professionals have raised concerns about several emerging risks:

Overreliance on AI during emotional distress
Development of unhealthy emotional attachments to chatbots
Reinforcement of harmful beliefs or delusions
Excessive validation without meaningful challenge
Reduced engagement with human support networks

These concerns become particularly important for adolescents and young adults, a stage of life where social relationships and emotional development play a critical role.

The Governance Challenge for AI Companies

The rapid adoption of AI in sensitive areas such as mental health raises an important question for technology providers: how much responsibility should AI companies bear for user outcomes?

Several AI vendors have already introduced safeguards designed to identify signs of self-harm, direct users toward crisis resources, and provide additional protections for younger audiences. At the same time, regulators in states such as California, New York, and Illinois have started introducing legislation aimed at reducing mental health risks associated with conversational AI.

The discussion reflects a broader trend that extends far beyond healthcare.

As AI systems become more integrated into everyday decision-making, organizations must think carefully about governance frameworks, safety mechanisms, transparency, human oversight, and risk management. These considerations are becoming essential components of responsible AI adoption.

What Businesses Can Learn

For organizations implementing AI-powered assistants, customer support agents, or internal knowledge systems, this research offers an important reminder: user trust grows faster than technology capabilities.

When people interact naturally with conversational AI, they often attribute greater understanding, empathy, and authority to the system than it actually possesses.

Building responsible AI solutions therefore requires more than technical performance. It requires clear boundaries, escalation mechanisms, human oversight, and thoughtful design choices that guide users toward the right resources when situations exceed the capabilities of automation.

As AI continues to evolve, the conversation is shifting from what these systems can do to what they should do. For technology leaders, software teams, and regulators alike, that distinction may become one of the most important challenges of the AI era.

Source